| { | |
| "_name_or_path": "google/vit-large-patch16-224", | |
| "architectures": [ | |
| "ViTForImageClassification" | |
| ], | |
| "attention_probs_dropout_prob": 0.0, | |
| "encoder_stride": 16, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.0, | |
| "hidden_size": 1024, | |
| "id2label": {"0": "bed", "1": "chair", "2":"couch", "3":"table", "4":"wardrobe", "5":"clock", "6":"keyboard", "7":"lamp", "8":"telephone", "9":"television"}, | |
| "image_size": 224, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 4096, | |
| "label2id": {"bed":"0", "chair":"1", "couch":"2", "table":"3", "wardrobe":"4", "clock":"5", "keyboard":"6", "lamp":"7", "telephone":"8", "television":"9"}, | |
| "layer_norm_eps": 1e-12, | |
| "model_type": "vit", | |
| "num_attention_heads": 16, | |
| "num_channels": 3, | |
| "num_hidden_layers": 24, | |
| "patch_size": 16, | |
| "problem_type": "single_label_classification", | |
| "qkv_bias": true, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.29.2" | |
| } | |